I’m excited to share with you a new tool We have been working on called tfrobot. It’s designed to help you quickly deploy many VMs on the ThreeFold Grid at once.
What can the tfrobot do?
- Deploy lots of VMs at same time: No more one-by-one struggles! Sometimes deployments might fail, but tfrobot will always try again for you!
- Cancel all VMs with one command
- Load all deployed VMs at once: See everything deployed by tfrobot in one place if you missed the logs in the terminal.
- It currently supports yaml and json for configuration and output (for further processing)
How to get tfrobot:
- Go to releases page in the GitHub repo releases: https://github.com/threefoldtech/tfgrid-sdk-go/releases.
- Unpack the download.
- Put the “tfrobot” file somewhere your computer can find, preferably in $PATH directory like /usr/local/bin
- Make a config file (example name: config.yaml). Tell tfrobot what nodes and VMs you want it to deploy. You can use some of the examples in the repository for inspiration https://github.com/threefoldtech/tfgrid-sdk-go/tree/development/tfrobot
- Now let’s deploy! Open the terminal and type
tfrobot deploy -c path/to/your/config.yaml. (Make sure to change the file path here before running)
-denables debug logs
P.S. Please be careful with storage sizes – put them in GB! It’s also possible you may end up using more nodes than you asked for because of those retries on failed deployments.
How tfrobot helps keep the Grid strong:
By making it easy to deploy lots of VMs, tfrobot helps put the whole Grid through its paces! We use it to test how things work under heavy load or try out experimental setups.
In essence, tfrobot has not only served as a tool for simplifying the deployment process for us but also as a catalyst for ongoing enhancement of the ThreeFold Grid. It has played a crucial role in testing the grid’s reliability and the continuous improvement of its components. Through the extensive testing of tfrobot, we’ve been able to simulate a wide range of deployment scenarios, from small-scale setups to large, complex configurations involving numerous VMs across different node groups. This comprehensive testing has highlighted areas for enhancement, leading to significant improvements in grid performance, stability, and usability.
Give tfrobot a try and let us know what you think. Questions are always welcome!
